Qualities of consent: an enactive approach to making better sense

Philosophical work on the concept of consent in the past few decades have got to grips with it as a rich notion. We are increasingly sensitive to consent not as a momentary, atomic, transactional thing, but as a complex idea admitting of various qualities and dimensions. In this paper we note that the recognition of this complexity demands a theoretical framework quite different to those presently extant, and we suggest that the enactive approach is one which offers significant value in this regard. In consonance with arguments made by Laurie Penny about how consent is a continuous and dynamic process, we outline how an enactive approach identifies consent as temporally extended (rather than momentarily transactional), and as affected by the skilfulness of the agents involved, the fitness of community-provided resources to negotiate the consensual relationship over time, and the unfolding of circumstances in the situation in which the joint action is taking place. We argue that the value of an enactive perspective on consent is in highlighting these complexities, and in providing resources to articulate and theorise them in ways that are not open to other current approaches.

In reflecting upon her own and others' consensual histories, the journalist and author Laurie Penny voices a concern of this sort."Technically," she writes, friends and acquaintances of hers may have just fallen on the right side of establishing consensuality in their social interactions (2022, p.49).This demonstrates that they have been able to negotiate the important "technicalities to be overcome" in order to achieve consent with others, which serve as a "baseline" for pursuits of "pleasure and desire" (ibid., 12).This 'baseline' may be a word or gesture that a judge, a social worker, or a friend may recognise as satisfying some minimal threshold for establishing consenting relations.However, Penny argues, the question is whether endeavours to reach consent, theirs and ours, ought to stop there.If we can grasp "the problem (…) that technically isn't good enough" in many a social interaction, i.e. that the forms of consent we might reach or be party to seem in some ways imperfect, there then arises a question of how one might strive for other, better types of consent with our social partners, and how we should best make sense of them.In other words, instead of understanding consent in terms of some minimal conditions to be met, the idea is "to rethink (…) consent" in terms of a dynamic framework that might distinguish between various realisations of consent, not all of which are created equal.These we might think of as differing 'qualities of consent.' There can be little dispute that contemporary philosophical discussions of consent have developed sophisticated analyses and concepts when it comes to the "technicalities" of consent.Moreover, a number of contemporary philosophers can be argued to have touched upon the issue of 'quality of consent.'Where this issue has been raised, however, the focus has been on the ways in which historical or circumstantial factors (e.g.ignorance, inebriation, misinformation, miscommunication) can have the effect of disrupting the state of consent in situations that otherwise appear to bear the hallmarks of "technically" consensual activity (cf Dempsey, 2021;Dougherty, 2013Dougherty, , 2018;;Hansson, 1998;Liberto, 2017Liberto, , 2021;;Mandava et al. 2012;Matey, 2021; see also 'consensual minimalism' and 'consent-plus' theories of consent, Wertheimer, 2003, p.131ff).
We believe the phenomenon highlighted by Penny invites a richer conception that implies a dynamism and temporality in consent, one that presents important challenges that have not yet seen much discussion.How might one ground and develop the idea that consent is a skilful, constantly renegotiated relationship?How should we consider it systematically, sensitively, and ethically as a fluid and potentially unstable (if not uncertain) phenomenon, subject to ebbs and flows in terms of its content and its normative implications for the persons involved?These things are hard to discern in the existing literature.In particular, little attention has yet been given to how qualities of consent come about in terms of how some forms of consent are nested and announced within others, in ways that may not always perfectly overlap.
While there are certainly tools provided by different philosophical approaches for these questions, for the sake of space, coherence, and focus, we shall explore them by broaching the premises and implications of an enactive account of consent.The enactive approach is attractive for this purpose because of a theoretical commitment to a continuously evolving, dialectical relationship between a person and their individual and social environment (Thompson, 2007;Di Paolo et al., 2018).The enactive approach has a tendency to transform dichotomies into continuities (McGann & De Jaegher, 2009), and it is precisely such a shift in viewpoint that we see Penny putting in the offing with regard to consent.
We will proceed as follows.First, we will briefly sketch the desiderata for a theory of consent highlighted in some contemporary discussions (Sect.2).On the basis of that sketch, we will give a preliminary justification for why an enactive account might offer an alternative notion of consent, and especially as one amenable to a concern for dynamic qualities of consent.In the next part of the paper (Sect.3), we will then turn to an enactive account of consent expliciter, along the way laying out some of the basic enactive concepts and principles which make it available.We follow this (in Sect.4) with some lessons for thinking about consent within the enactive account and point to a key question for its further elaboration.

Philosophical knots of consent
Before anything else, what do we mean by consent?Here we shall spell out its main facets in relying upon an influential body of literature situated somewhere between the philosophy of law, meta-ethics, and applied ethics.At the risk of oversimplification, it will be useful to focus on the following case of consent, which can be taken as broadly representative of a number of important questions pursued by consent theorists today.
Lawn Blower.A asks B whether he can borrow B's lawn blower, about which (unbeknownst to A) B is quite proprietary.B mistakenly thinks that A has asked to borrow his lawn mower, about which he is not proprietary.B says, ''Sure, it's in my garage; help yourself when you need it.''(Miller & Wertheimer, 2010, pp. 85-86) Is there a form of consent realised here, and if so, how?What are its implications?Roughly following Liberto's presentation (2017), we see certain contemporary discussions as tending to focus largely on two types of issues with respect to cases like this: what 'the consent' is, and what it implies for agency.
The first technically-demanding issue has to do with the 'ontology of consent' (Alexander, 2014), i.e. the establishment of the "validity of the consent" (Matey, 2021).This question concerns the requirements or conditions of consent, as grounded in factors such as (expressive) action, knowledge, intention, and voluntariness on the parts of consent participants.In terms of the ontology of consent, the above case nicely demonstrates one such central issue; namely, what it is that makes the consent 'real' or gives it traction as an obtaining state of affairs.The underlying premise in such a question is that at any given time a state obtains such that consent exists or does not; more controversial is how to account for the grounding of that state.There is debate here on whether consent is a "behavioural" or "mental" phenomenon (Dougherty, 2019, 1-2), that is, whether it comes about through some kind of act that requires overt, expressive performance, if not convention, which may or may not be jointly achieved.For instance, under a closely related cousin of the so-called 'behavioural' model, namely, a performative model,1 the obtaining or reality of consent comes about through a one-time [and thus temporally isolatable], concrete [thus materially determinable], and felicitous [thus consciously intended, perhaps conventionally prescribed] exchange of a communicative (oft linguistic) token between social partners.By contrast, others believe that consent 'merely' requires one's having of a certain mental state, for instance, of acceptance (Alexander, 2014;Reinach, 2013) or perhaps a certain desire or interest (Hume, 1975).In essence, this discussion comes down to whether, in the above example, B would actually need to utter the words 'help yourself,' or whether it would already suffice for B to want or accept that A would make use of the mower.
The second complex issue has to do with the grounds for wrongs in both consent and non-consent states of social interaction -in short, the normative implications of consent.Again, this issue fractures into a number of questions, but two of its most important, related facets may be said to be those of trespass and normative transformation.On the one hand, questions of trespass are about understanding the kind of wrong involved when someone violates a state of consent or when one acts without consent.These two senses can be thought of intra-and extra-consensual forms of trespass.In terms of the latter, many but not all forms of trespass are wrongs; as Liberto points out, some trespass may be justified when "I may permissibly walk on your property if I need to do so in order to save a person lying injured on your land."(2017,129).Yet establishing the line of demarcation between justified and unjustified trespass is no small challenge.The former form ('intra-consensual') of trespass raises its own sort of questions; for instance, when consent has been established and when (personal) boundaries (or rights) have been violated, is this wrong a wrong that is in some sense distinct from, e.g., the bodily harm through which it might be effected?This may be harder to conceive of (but not impossible) in terms of lawn mowers and lawn blowers, but in both intimate and medical situations anyone having the sense of undergoing something they have not 'signed up for' will readily recognise the stakes of this kind of trespass.In cases of violations of consent which do not involve harm, theorists have investigated how to understand the 'wrong' -where it comes from and why -when there is no harm.This is a question analogous to the question of the kind of wrong it is to break a promise. 2  These last considerations point to another crucial facet of consent, where normative transformation has to do with the ways in which certain interactions (bodily, emotional, cognitive, social) become enabled under the special conditions of consent, when they might otherwise not be. 3This notion too breaks down into different senses, one more 'objectively' anchored, and another more 'subjectively' or agen-tially focused.The former is often considered in light of how consent seems to alter the nature of certain apparent wrongs; some things, like taking a lawn mower from a neighbour's garage or touching someone, are wrong without consent, which then become allowable or even expected consensually.Not every sort of wrong seems amenable to being affected by consent in this way -a parent's consent that their child work all day in a factory still seems wrong, despite any consent. 4Thus, if some things may always still be wrong, even with consent, then there are challenges in understanding which sorts of wrongs can and cannot be affected by consent, and why (e.g.how there could be consent to certain forms of bodily mutilation but not to be murdered).
The other, 'agential' side of the question of normative transformation has to do with the special powers that consent seems to bring about in persons; powers to make the out-of-bounds or the forbidden available, appropriate, or acceptable. 5Consent seems able to change the moral or normative status of certain forms of action and thus seems to enable certain specific form(s) of agency, meaning the ways in which one partner or constituent of the consent may or may not act with respect to one or more corresponding partners in the consent relation.With our consent, our garage or our body becomes an available field of action for the consentee, in a way that it was not prior to our consent.In other words, consent can be said to involve such transformation insofar as it makes possible actions or attitudes by one partner, towards another partner or their interests, which without the obtaining of the consent would have been impermissible.For instance, when consent is conceived as a transaction according to the performative model (Bolinger, 2019;Manson, 2016;Beyleveld & Brownsword, 2007),6 its obtaining entails that certain rights and corresponding duties are instituted or created; these empower social partners to act in certain ways with respect to each other (but only in those ways!) and also to evaluate each other with respect to either appropriate actions taken or attitudes held in view of the 'social fact' established by the exchange of the consent.

Continuities and competencies of consent
Any enactive account of consent would ultimately have to be answerable to all these different facets of current debates on consent, but it is in terms of this latter aspect in particular that we can make clear both what "qualities of consent" might be about, and what an enactive perspective might bring to that discussion.Before we outline the enactive account as we see it, we will take a little time to unpack more carefully the notion of transformed agency just invoked.
To recap, the agential sense of normative transformation has to do with how a distinct form of agency comes about through consent, one that had not existed prior to it and which could not exist without it.Importantly, once such agency has been established through consent, what matters as much as any specific action is the consenter's exposure to a range of potential actions by the consentee (we note that in many or most situations these roles are mutual and reciprocal -that consent goes both ways).Some of them may be reasonably anticipated by the consenter, but others may not.For example, it would not be reasonable for someone to expect to know beforehand all the possible ways in which their neighbour might use their lawn mower.Moreover, only some but not all of them might seem problematic or verging on trespass to the consenter, and yet, in granting consent, one is signing up for that open-endedness of the neighbour 'helping themselves' to the lawn mower.Likewise, for someone consenting to take part in medical research (or in intimate relations with another), not only will one have to be open to undergoing certain specific concrete actions, like getting a shot, but also open to, and in some sense living in uncertainty with, a set of possible interventions that may be taken by the consentee -like being made to wear an unflattering gown or being touched in some unanticipated manner.This open-endedness, however, far from implying some kind of expansive commitment to all that follows, brings with it a responsibility for a continuous judicious sensitivity to the consent relation itself. 7n important counterpart to agential normative transformation (on the part of the consentee) can therefore be recognised in a transformed normative exposure on the part of the consenter, and what seems crucial about the latter is that it is defined by an uncertainty, if not outright precarity for one or both of the social partners in consent.
To put the same point differently, consent is not about the eradication of risk in social interaction, but rather about a conversion of one set of risks in social interaction into another.Consent is less an acquiescence to all of the implications and inherent risks of a given situation, but rather an agreement to mutually manage the particular risks within that situation.
It is crucial to see that being exposed to these risks and managing them in the social interaction can play out in different ways.In reference to a medical context once more, having someone who has given medical consent wear a certain gown or unexpectedly touching them with something cold may not 'break' the consent but may still make that consenting person feel quite uneasy.In that thought one has a strong intuitive basis for understanding why the 'quality of consensual interaction' can be compromised, and this compromising will have important implications for the further trajectory of this social interaction, if not in others as well.However, the quality of consent must not only be seen as able to be compromised or downgraded.The ways risks are managed through consensual interaction may no less open up new avenues for meaningful interaction with others; the manner in which we are sensitive to our neighbour's exposure to our agency, when they consent to our taking something from their garage, may enable and engender further consensual relations initially unavailable to us.That is, consensual relations may be subject to differing qualities of realisation in respect to each other, namely, in terms of the uncertainty and risk incurred through the consent, and the manner in which living with uncertainty is managed and coped with throughout the social interaction.Now, if the question of 'qualities of consent' is framed in such terms, i.e. variations of normative exposure to uncertainty and risk for consenter and consentee within and across forms of consent, there is little such discussion of 'qualities of consent' in the consent literature.Typically, the normative or "morally transformative" power of consent is taken to be co-extensive with, and non-varying within, the establishment of "valid consent."8This means that, within such discussions, a timeline of when an agent has access to one or other normative or 'morally transformative' agency perfectly maps onto a timeline of when minimal or threshold conditions for 'valid consent' relevant to that agency obtain.With the idea of qualities of consent we have just sketched, we are suggesting something quite different; those two timelines do not perfectly line up, indeed never do, due to how variation in agency and exposure affect how the consent is played out.Our indifference or sensitivity to the patient's discomfort at wearing the gown affects in realtime the normatively-transformed agency that we have access to through their consent; thus neither access to such agency on our parts, nor exposure on the part of the patient, is a binary or on/off state.
There may be some important reasons for why 'qualities of consent' on our construal have been little discussed, having to do with the quite challenging nature of such an issue.Here is a way to think about those challenges: nothing a priori about the claim that forms of consent scale up and down, and that some forms of consent may be nested within and enabled by others, is opposed to the idea that consent may be a real, actionable fact about the world.However, from the moment one claims that there are consents, each again with a distinctive quality, and not just consent singular, there can arise a hesitation about whether there are different obtaining conditions or norms of enforcement relevant to each; one might then well wish to know how these stand in relation to each other.Are some of these 'qualities of consent' more real than others, for instance, as where our consent to being touched by person a would obtain differently than our being touched by person b? Are some more demanding or freeing, or indeed, better or less desirable than other forms, and how might one account for such variation?
We are not suggesting that extant theories lack any means to answer such questions.Some available avenues here would be Liberto's explorations of the role of risk in consensual (2017, 2021), Bolinger's notion of "normative opacity" (2019), discussions of 'consensual minimalism' (Wertheimer, 2003), and, as also mentioned, the extensive analyses of situations (like "Lawn Blower") that seem to stretch the meaning of consent to its breaking point.However, evaluating the sufficiency of such accounts is not within our current scope.Instead, we are interested in what an enac-tive account would bring to the issues raised by "qualities of consent."On our view, there are a number of reasons why one might turn to an enactive approach, at least to evaluate the resources that it might dispose of to treat them.For the following reasons, we can motivate that turn on the basis of the sketch of 'qualities of consent' just given, without yet furnishing a fuller picture of enactivism.
A first reason is the enactive approach's recognition of the temporally extended character of human interactions (De Jaegher & Di Paolo, 2007;Di Paolo et al., 2018;McGann 2014;Thompson, 2007).These are not momentary phenomena but are indeed constituted diachronically (Kirchhoff, 2015) through the enactment of a number of related skills.From the point of view that our behaviours and social interaction are constantly evolving from one moment to the next, consent too must be seen as a fluid human interaction subject to various forms of processual realisation in view of its context and participants -for instance, where consensual interaction between persons might (or might not) manifest qualities of type a, then type b, then type c, e.g. of uncertainty and openness and their navigability, etc.
Taking this thought further, one crucial implication here is that no token or instance of consent occurs in a vacuum; no form of consent stands on its own.One quality (or quality-type) of consent is embedded within others, and it is always possible that the quality of consent can degrade or dissipate as well.
This may seem like a difficult thought to think; one's consent is typically considered to be something standalone.We consented for you to borrow our lawnmower but not our Ferrari, and for good reason; it's just the lawnmower that we want our consent to pertain to, because it's just with regard to our lawnmower that we're open to being exposed to your agency.Yet it would be unreasonable to think that our lending you the lawnmower would have nothing to do with our (potentially) lending you our Ferrari, even while the one form of consent in no wise guarantees the other.What is thus challenging but fascinating about this 'processual' picture of consent is that it places front and centre the claim that consent may not only be a changing outcome of human interaction, but also an unstable and open-ended (though not arbitrary) one.The enactive perspective is certainly not the only domain within the human sciences to embrace an understanding of human agency as uncertain and open-ended.However, it is particularly well-placed to explore the radical implications of precarity and instability for the human organism and its social relations.An enactive approach would thereby be an advantageous route to analyse how there may be both continuities and disconnections between qualities of consent, via the continuous and yet distinct processes of human interaction by which they would be realised.
A second consideration here has to do with how competences should factor into the coming about of agential normative transformation via consent.In contemporary discussions, the import of competence 'in and of' the consent often seems set aside from the outset.Competence has sometimes been taken to fall under situational or constitutive conditions outside an agent's immediate control in the situation of consent, for instance in terms of factors like intoxication, duress, knowledge, or maturity, which may or may not affect whether the consent is reached or given.It is in this sense one can find some reference to 'quality of consent,' meaning the competence of persons at the moment that they would be given or issue a form of consent, for instance as in Matey's claim (2021, 381) that "[b]oth the validity of the consent and quality of the consent can be compromised in different ways."9However, on such accounts, consent competences are not taken to play much of a role in the interaction of individuals in the consenting relationship, or if they do so, only negatively, by eliminating consent as a possibility.
By contrast, on the view that consent may be broken down into differing qualities, the achievements of which are highly contextual and mutually dependent on each other, competence would have to figure as a meaningful if not indispensable component of consent in its varying achievements.The competence of the agents involved in the consent may be a key way in which qualities of consent are distinguished. 10In addition, if qualities of consent can scale up and down in some non-arbitrary fashions, and perhaps according to different dimensions -in the sense that one form of consent can transform into and enable another, however fraught that process might be -then an appealing candidate for understanding this process is to consider how certain forms of consent foster or engender competences for further forms of consent.11This basic sketch of 'qualities of consent' provides a strong justification for an interest in an enactive account of consent, which takes as a starting premise the idea that forms of life are defined by the skills and competences by which they deal with tensions within themselves, and between themselves and their enabling environments.Skills and competences are in other words the basic grist of the enactive mill, when it comes to understanding the many facets of what it means to have embodied existence and to share a world with others.
We do not claim that only an enactive approach can provide conceptual resources to address these questions.Indeed, contemporary embodied cognitive science includes a bolus of perspectives, which sometimes seem consistent with one another, sometimes not, sharing some aspects of the approach presented here.We will not presume to short circuit the discussions around the commensurability of these points of view (e.g.Di Paolo et al., 2022;Fultot et al., 2016;Heft, 2020;Villalobos & Dewhurst, 2017) and claim that apparent inconsistencies are illusory, or that only one perspective as it stands now will have all of the answers.Our intention, as stated above, is to move forward with the resources of an enactive approach.

Enacting enactive consent
If open-ended exposure to another's agency are the risks intrinsic to the consent interaction, and if such normative exposures are not to be seen as discontinuous from each other within and across instances of consent, how should we best account for this?It is to such questions that we shall now turn, in providing a sketch of an enactive account of consent.While the approach does not offer off-the-shelf solutions to all of these challenges to our understanding of consent, we aim to demonstrate that it furnishes promising resources for properly getting to grips with them.In offering this sketch, we will have to undertake a few steps; first, a characterisation of sentient agency such as to lead to an account of specifically (human) linguistic agency; second, an understanding of the skill(s) with which that agency is deployed; and third, a depiction of social interaction, from which we can draw some considerations about consent from an enactive perspective.

Normativity in agency
By way of introduction, the enactive approach -not as a monolithic theory, but rather a group of basic standpoints currently being developed -is substantially characterised by a commitment to continuity.12An important manner in which this commitment is exemplified is in its understanding of agency -the potential for action -as it pertains not just to humans but more broadly to sentient organisms.
From an enactive perspective, agency is a continuous process of maintaining coordination between the precarious norms of the animated agent, and the complex flow of environmental events arising for that agent.Enactive researchers in other words hold that any agent is animated by insuperable needs for environmental coordination.This need for coordination grants a base normativity to that agent's activities -a bare measure of success and failure which exists not in the absolute binary of continued existence or dissolution, but in the vector of movement away from or toward destruction (Barandiaran & Egbert, 2013;Di Paolo, 2005).Enactive norms are thus instantiated by networks of productive processes from which one or other action may emerge.These networks are likewise precarious, in the sense that they are both continuous and interruptible; there is no permanent stable point that such networks can reach, from which their continued existence is guaranteed.As such, any living system cannot be continuously, stably safe; actions only mitigate or transform rather than completely remove the tension between the agent's existence and the impersonal flow of dynamics of the environment in which it is embedded.Until the system is destroyed its activities are incomplete and open-ended, radically partial, though it is also certainly true that some states are less perilous or fraught than others.
The basic conceit of this enactive notion of agency is that agents are systems organised so as to support the continuity of precarious processes of environmentaland self-coordination.The range of behaviour of an agent is the range of actions it can take in the interests of maintaining itself with respect to its norms of continuity, theorised as self-sustaining or autonomous networks of activity at a given level of description.Autonomous networks are those sets of processes within a medium which have a mutually supportive relationship, such that any given process in the network supports the continuation of at least one other process in the network (Varela, 1979).This organisation is a dynamic thing, though it also has structural (that is, particular material, spatial) aspects.A particular autonomous system will be instantiated in a given set of material processes -with all of the texture and complexity that chemical, biological, and in some cases social, processes have to offer.This physical instantiation is the network's body, and the manner in which its body both enables and constrains the possibilities for the network is its embodiment.
The paradigmatic case of agency and normativity for enactive researchers is therefore the organic body.For enactive accounts, it has been a substantial undertaking to recognise, and develop theories to properly account for, richer forms of agency that proceed from that paradigm.In Linguistic Bodies (2018) Di Paolo, Cuffari, and De Jaegher (following an initial framework broached in Varela et al., 1991) distinguish between organic agency, sensorimotor agency, and linguistic agency.At the base is organic agency (with concomitant organic bodies), the fundamental basis of life and mind, grounded in biological processes of self-production.Sensorimotor agency (and its associated sensorimotor bodies; Di Paolo et al., 2018) arises in a richer domain of activities in which the autonomous systems in question are not organic, but activitybased -here one can think of how habits and skills emerge as systems of environmental coordination.However, while this may seem to encourage the idea that, according to enactive thinkers, one kind of body or agency is built on the foundations of another, they are equally at pains to stress that sensorimotor agency, and sensorimotor bodies, are not simply built 'on top' of organic bodies.Sensorimotor agency becomes part of the medium in which the organic bodies operate.Though organic norms impose certain tendencies and constraints on sensorimotor activity, the sensorimotor similarly gives rise to new pressures and constraints for the organic too, transforming organic bodies rather than just layering over them.Sensorimotor bodies are therefore inherently complex, involving not just multiple processes of production (necessary for the simplest forms of environmental coordination according to enactivists' use of Varela's principles of autonomy) but multiple norms, which are inevitably in tension with one another.Sensorimotor life is a continuous (and never perfectible) process of keeping these various norms in coordination with one another through different forms of agency.
The same logic then applies to the domain of linguistic agency and linguistic bodies.In this case, through a complex of processes involving multiple agents over prolonged time, a stable domain of agency, and of appropriately cultivated bodies, emerges.While the linguistic domain may seem relatively stable, for instance in having consistent community practices, or a readily identifiable and reasonably reliable grammar, the agent within it is not.Just as sensorimotor agency involves the continuous coping with the pull and push of multiple norms of habit, skill, and bodily need, so too is linguistic agency a process of coping with a complex normativity which exists not in, but through various sensorimotor bodies.Just as sensorimotor agency transforms the organic bodies through which they arise, so too does linguistic agency shape the sensorimotor dynamics in which it emerges.Human beings, as linguistic agents, thus act within fields of force that are not a simple combination of multiple individual actors.Rather, networks of autonomous processes, instantiating different kinds of norms, arise within groups, communities, and societies, giving rise to new forms of agency and bodies.Though linguistic activity transforms the sensorimotor and organic bodies that give rise to it, the ebb and flow of organic and sensorimotor activities which give rise to the linguistic domain in the first place still matter.As such, linguistic bodies are individual things; there is no single, standard, linguistic body, but rather a plethora (billions) of linguistic bodies (Di Paolo et al., 2018), producing a dynamic, effervescent medium of activity within which those linguistic bodies are in a constant process of emerging: always becoming, but never complete.

Agency in context
So much for a rough and ready presentation of the enactive account of actions and human actors qua embodied linguistic agency; the next step is to see how this understanding of agency leads to a notion of skill in agency, and specifically of skilful participation in embodied life.The very notion of skill arises when we note that the enactive approach to agency entails that there are a range of complementarities and mutual relationships between the different aspects of what it means to be a living, sentient organism (i.e.organic, sensorimotor, social).The agent and its environment are mutually specified, each implying and only to be understood in terms of the other.Similarly, perception and action are mutually specified, such that it is our capacities to act (arising from our agency) that determine that to which we will be sensitive.Here, think of how learning to play a musical instrument makes us sensitive to aspects of music, rhythm, and performance to which we would otherwise be oblivious.Our sensitivity to different aspects of the world around us emerges with our capacity to act in that world.
In the case of social interaction, it is through such skill-structured sensitivities that we come into contact with other human beings.Our sensitivities to other people are structured by our skilful participation in social activities.Contact with other human beings is achieved not just by the co-presence of organic bodies, but also by the capacity of mutual influence and coordination that emerges in the enactment of various social skills.We come into contact with others in contexts in which we are capable of interacting with them, engaging with them within the flows of particular modes of activity. 13 Despite the broad recognition of the complexity of social agency, the case of agent-agent interactions would seem to be a particularly complex and difficult challenge for the enactive perspective.It can be tricky enough for one agent to cope with 13 We note here that there are a range of related embodied perspectives on social activity where similar insights are made available.Chemero (2016), for instance, proposes that social interactions are characterised by synergies that arise in the interactions between multiple agents, where individual activities can be characterised as the product of a single system that nevertheless involves multiple people.Crippen (2022) and Crippen and Rolla (2022) relatedly examine how emotional expressions are products not just of processes that run outward from a single person's physiology through their face and limbs into the environment, but arise in the environment and are coloured by an active context.the environment alone; how much more complex does it get when two agents must coordinate not only with the environment, but with each other and other agents as well?In reality, of course, these different aspects of the coordination problem are not independent of one another, and where structures allow, the more complex problem can start to become part of its own solution.For instance, we tend to encounter others within the process of on-going environmental interactions.Typically, our shared environment, and substantially shared embodiments, serve to orient multiple agents such that, at a low level of granularity, relative coordination is already in train as the agents encounter one another.
Our cultures and communities organise physical environments and social practices in such a way as to ensure a relatively high degree of what we might consider baseline coordination even before two people have met -without such baseline coordinative structures, the people are in fact quite unlikely to meet.We can illustrate this idea with a brief look at two such coordinative structures that have been discussed.
The first is what Roger Barker (1968), Paul Schoggen (1989), and others have termed a "behaviour setting."A behaviour setting is a standing pattern of behaviour with a particular temporal, physical, and social organisation.Our lives are spent in behaviour settings of various kinds, from a language class, to a religious service, to a seminar, to a working retail venue.Behaviour settings are organised in multiple ways, having temporal structure (they begin and end, and often provide for specific activities in a given sequence), they have physical structure (they typically involve specifically designed and furnished physical environments, such as a class room or temple), and social structure (they provide roles, with different valid actions available to those roles).Getting most things in our lives done means bringing our behaviour into coordination with the set of constraints and resources provided by the behaviour setting in which we find ourselves (though in most cases this provides only broad guidance, with a wide range of specific actions within the setting still possible).
The second means of ensuring that two people are already in some relative coordination once they come into contact with one another is Di Paolo, Cuffari, and De Jaegher's (2018) concept of a "participation genre" (a term they derive from the work of Bakhtin, 1986).Related but distinct from behaviour settings, participation genres are somewhat more broadly defined and characterise the various kinds of interaction in which a person might become skilled at engaging (and we might see different genres being made available with different degrees of salience within different cultures).Examples offered by Di Paolo et al. include collaborative work and team sports, which work within a broader set of "life genres" such as gardening, commuting, cooking, and in turn organise "gesture genres" or "speech genres" such as phone conversations or poetry readings.
The enactive claim here is that practices and places, instantiating genres and settings, substantially shape people's capacity to manage themselves and each other, and to engage in shared actions.The actions, activities, and expressions of individuals are enacted and encountered in the context of already shared activities which plays a significant role in giving them shape and meaning (Crippen, 2022;Crippen and Rolla, 2022;McGann, 2015;Di Paolo et al., 2018).
Here we see the contour of an enactive approach to the earlier discussed issue (Sect.2) of the obtaining or 'reality' of consent.Genres and settings create comple-mentarities in the actions, qua different moves and movements, of participants.It is thus in them, as much as any discrete actions within them, that consent is grounded.My offering of money only makes sense because of the retail setting I am in, and through the complementary role of vendor played by the cashier I am speaking with, for instance.Likewise in consent situations, roles in genres and settings can be realised with greater and lesser degrees of skilful participation.One of the key ways in which such skill is demonstrated is the sensitivity participants must show to the other person or persons with whom they are engaged.This skill can be significantly evinced in at least two ways.The one is in terms of drawing on repertoires of genre-and setting-relevant behaviours (some creative, some conventional in their application; some awkward, some more reassuring).Yet this skill is also shown more personally, in terms of how the one agent is sensitive to how the other faces the uncertainty and ambiguity of the normative exposure entailed by available moves within those (applied) genres and settings, as when one is asked to show the contents of one's bag, or when one becomes the target of a crunching tackle on the football pitch (cf.Owens, 2012).
The part played by communities and cultures in such skilful participation is crucial, insofar as the genres and settings made available within a given culture or community instantiate the norms of that community.Thus, in view of the repertoires available to agents and the (communal) norms embodied in them, some actionsmoves and movements in coordinating with others and in remaining sensitive to tensions with the interaction and in others -will be made salient and easy, others obscure and difficult, depending on the resources that practices and places make available.Insofar as we are capable of engaging in particular participation genres, we encounter other people already within those genres, with roles to be performed, particular sets of actions to be considered more legitimate or likely than others, and certain finegrained details of action made available for enactment.

Social inter-enaction
What sort of picture of human social interaction emerges from the preceding, and what implications does it have for understanding consent?Within Di Paolo et al's (2018) approach, social interaction comes in a wide range of forms.The most basic is considered that of participatory sense-making, which is the form of interaction in which the interaction itself is an autonomous network that incorporates the agents interacting.Vital to the whole notion of participatory sense-making is that the interactors are in a situation of mutual influence, such that they engage in a process of mutual coordination.Though analogies are sometimes treacherous, we might imagine the formation of a vortex within a sea with multiple currents of different strengths.The different currents within the sea can affect one another in a number of ways.The force of one current pulls the water away from one flow toward another, affecting how the sea as a whole in that general area is moving.The sea is a complex of roiling (though not necessarily turbulent or chaotic) flows.Imagine now the interaction between some of these currents, and perhaps some quirk of the sea bed, shores, and landmasses, leading to the formation of a vortex or whirlpool.The vortex distorts the entirety of the fluid medium in its vicinity, while simultaneously being affected by a range of these currents within the sea as a whole.
In our analogy, the sea is the complex medium of human life, its edges and bed the various aspects of our material environment, the tides and currents the tendencies of societies, communities, families, and more -languages, practices, routines, and rituals, participation genres -operating at various temporal and spatial scales.These flows, instantiated in a specific vortex, are concretised as the habits and skills of an individual living person.The organisation of the individual (rather than, say, the specific water or even particular currents that form it at any given moment) is what matters, and what provides some continuity over time.Activity flows through and is affected by the individual, rather than being something that is encompassed and exhaustively explained by the individual.Now, whatever semblance of respect for physics we have attempted to hold up to this point must be cast aside, as we imagine multiple such vortices coming into interaction with one another.Because of the existence of tides and bigger currents in and through which the individual vortices arise, the meeting of such individuals can itself come to form a larger vortex -which encompasses, but crucially does not destroy, the original ones.In this case, this larger vortex creates constraints on the ways in which the individual vortices will behave -how they will move with respect to one another and the world around them.This is our analogy for participatory sense-making, where the interaction itself takes on an autonomy which constrains, but does not eliminate, the autonomy of the individual participants.Participatory sense-making does not only constrain, however, as it can also support activities of which neither individual agent alone is capable.To continue our analogy, the larger vortex is robust enough to persist in rough seas or shifts in currents that would have turned or destroyed the smaller ones.
This analogy of vortices large and small in constant interaction is meant to capture the key enactive idea that participatory sense-making, and the actions and interactions it includes, is driven by systems of precarious norms.Some of these are inherent within individual organisms (organic, living bodies), some are in the stability of organisms interacting with their environments (sensorimotor, skilful bodies), and some are inherent within communities of such organisms, and stabilised in their interactions (linguistic, languaging bodies).The messiness of this complex medium of activity, the near impossibility of all these various norms in a given system being coherent, means that being a sensorimotor, and certainly a linguistic body, is never about finding a single reliable balance or equilibrium point, any more than one might find a stationary body of water in a heaving ocean.Rather, agency is a continuous process of wayfaring through the shifting landscape of needs.To stretch our analogy even further, successful action is less a matter of navigating the sea, but more like continually surfing, riding the currents, moving through the medium by going with one set of flows, then allowing the set to change, flow by flow, according to what is possible within our immediate environment.
When actions, whether individual or shared, are animated and guided by precarious norms, interaction between agents is therefore normatively loaded even before two people actually encounter one another.Encounters, particularly where the agency and autonomy of the other is recognised, always involve some form of exposure and vulnerability (though the texture and character of the interaction, conditioned by available participation genres and other resources, will affect just how vulnerable the different interactors are) (Di Paolo et al., 2018;Fourlas & Cuffari, 2022).This enactive notion of precarious norms thereby offers both a detailed, dynamic description of the 'levels' of normative exposures to the agencies of others that agents pass through, and as well as a specification of the diverse 'risks' agents encounter through and across those different levels.In this way, the enactive approach takes ethical action, at one side of the spectrum, to be on a continuum with the precarious norms and tensions navigated in action more generally -wayfaring within a complex shifting landscape rather than navigating by some fixed and dependable principles.
Social interaction is therefore a complex and textured thing over several scales, and understanding it must be done in a cautious and sensitive manner.The particular set of skills that I have to engage with another human being will make me sensitive to certain aspects of that other person, but the extent to which those skills will apply will depend on how that other person enacts themselves.Where we share many such component processes (e.g.behaviour settings which we can both inhabit and participate in), coordination will be possible because we will be mutually sensitive.Where there is less overlap, coordination will be much more difficult and challenging, requiring, in essence, the construction of new, idiosyncratic participation genres and social skills from scratch or else the adaptation of previously employed ones.
Even where component processes are shared, limitations apply.If one's participation genres are systemically racist (as most are), then one's capacity -for instance as privileged, white, or male with limited social circles -to interact with people of marginalised backgrounds is impoverished, usually in ways to which that person is oblivious and which place asymmetric demands on others.Social interaction is thus not all or nothing, and is enriched or impoverished by both the extent of the medium we share (the number of "currents" we have in common), and their structure.

Consensual lessons
For the purposes of elaborating an account of consent, one can already anticipate, on the basis of the enactive perspective we've sketched, that there won't be any absolutes here, and also that any human social interaction is already normatively loaded even before anyone gets to the point of explicitly negotiating anything.But beyond these general indications, we may also have the resources to address the question of consent in interesting ways.
First, the enactive perspective discourages the view that consent be viewed as a transaction with discrete stages of fulfilment.Continuous, and dynamic, it is perhaps a participation genre that operates as part of other participation genres.It is more adverb than noun.We might better talk of "acting consentingly," rather than "giving consent."Second, an adverbial notion of consent means that different genres of collaboration and coordination will have different modes and manners of consent.Consent is contextual.In some cases, it is almost entirely informal and implicit in the form of engaging in a shared action or behaviour setting.In other cases, it is a formalised process, with paperwork, signatures, and other ritualised aspects.
From this point of view, then, consent occurs as part of joint action.Instead of involving some shared representation of a goal state, there is coordination of norms and intentions dynamically, responsively -a process of wayfaring together.Participation genres and behaviour settings (standing patterns of social activity within appropriately structured material environments) provide some structure, scaffolding in the form of typically occurring roles and responsibilities, and physical tools or furniture to constrain the participants and support that coordination.
As such, there are three key factors affecting consent from an enactive perspective.
The first is the skilfulness and sensitivity of the participants involved.The capability of each agent to perceive and respond to the actions of their co-participants is vital for effective and adaptive coordination between them.This includes the participants' capability to effectively use or incorporate the various forms of resources provided by their communities (participation genres, behaviour settings, and so forth).This means that there is not one skill or competence of consent, but many.
The second factor is the fitness of community-provided resources for the particular circumstances of the participants involved.These are resources to manage and cope with the normative exposures to other agents realised in consent relations, and as we have already noted, these resources can vary widely in terms of the sensitivities they afford to partners in social interaction.Many genres include systemic biases in roles and responsibilities that are likely to produce dramatic conflicts between the norms to which the individual interactors may be answerable and the norms inherent in the participation genre or behaviour setting.
Finally, the unfolding of circumstances over time affects the extent to which various resources become available or are invalidated.Skills or resources deployed under circumstances that apply initially can be undermined by the progress of matters which might otherwise be considered external to the interaction itself.A brute example of this might be an earthquake that rocks a research laboratory, necessitating that an experiment with consenting participants stops.In such circumstances there will be a taking stock, a careful consideration and most likely a review and reaffirmation of consent to continue.Subtler examples are more common.In recent years a change in the law in Europe -the enacting of the General Data Protection Regulation (GDPR) -has meant that data collected and stored under one legal regime have had to re-assessed, and in many cases destroyed, because the processes of consent that were applied when the data were first produced were no longer adequate.
Of these three factors, the first two are of most interest to us, as it is there that we see the emergence of tensions, conflicts between norms, and their resolution through coordinations.Where such coordinations are effectively managed, the actions are carried out consensually, but where such tensions arise and are not dealt with -whether because they are not acknowledged or because the complexity of the dynamics does not enable their resolution before some critical event -then consent lapses or is violated. 14or actions to be effected consensually, then, we should expect to see two things.The first is the resources, within the participation genres, behaviour settings, language or other communal resources in question, to support complementary roles, including appropriate means to adopt, negotiate, or shed those roles as required.We must also see a sufficient degree of skill in the use of these resources by the participants (this may vary by activity, some being more accommodating than others, some stricter in their requirements).The shared participation resources, and the participants' capability in deploying them, will make a participant sensitive to the other to some greater or lesser extent, part of which includes that other person's capacity to engage in the shared task and their commitment in doing so.The enactive approach has a strong tendency to make gradients of categories often thought dichotomous (McGann & De Jaegher, 2009), and it would seem that the case of consent is no different.Rather than a question of specific and clear boundaries which at any given moment may or may not be judged to be transgressed, consent is, as Laurie Penny (2017) has also observed, more like giving someone your attention -a continuous process.At any given point during the shared activity in question a person may be becoming less certain of their willingness to participate, or more.This may occur very slowly, but can obviously happen suddenly too.
Concerning the idea of 'qualities of consent', as we have shown, the enactive approach typically adopts a dynamic or processual conception of phenomena.In Linguistic Bodies we see the basis for a similar perspective with regards to consent.Consent is something that is managed in an on-going manner over time rather than transacted and fixed at a specific moment (even in cases where there might appear to be actions, procedures, or rituals which have the character of such transactions).
Our point here is not simply that consent comes in degrees -thus 'qualities of consent' does not just come down to how someone might be more or less consenting.The dynamic processes inherent in the enactive model go further than this.Linguistic interaction, from this perspective, is always a precarious balance of forces that, while they may reach brief or even relatively prolonged periods of stability, are never wholly fixed.As such, consent is a vector rather than just scalar term.A scalar conception would allow for the identification of the presence or absence of consent, or even indicate a point on a gradient of consent from whole-hearted commitment to vehement opposition.But the enactive model suggests that at any time there is, in addition to any such indication of state, some indication of change.As the interaction between the agents is being regulated through the enactment of a suite of actions, of moves that nuance or finesse the shared activity so as to address the layers of dialectical tensions, so too consent is continually improving or dis-improving.
Di Paolo, Cuffari, and De Jaegher (2018) repeatedly affirm (p.114, p.151, p.178) that dialectical tensions that are inherent within linguistic interaction cannot be addressed in a complete manner.There is no universally stable "golden mean", but rather a continuously shifting complex of forces that participants must manage using the resources their linguistic community makes available.This radical incompletecoordination,' it is on our view unclear how such criticisms apply to it.For a discussion of some problems associated with the "social coordination hypothesis" in the cases of promises, cf.Owens (2012, p. 136 ff.), Reinach 2013.ness, the impossibility of an exhaustive satisfaction of the tensions in the interaction, ensure that all participants are always regulating the encounter, tuning their own behaviour and influencing that of the other(s).At any given moment the encounter can be read as either heading toward consummation or toward dissolution.In the primordial tension between individual and interactive norms, consummation is the case where coordination is maintained such that the interactive norms ebb and individual norms come to dominate without the coordination between the participants breaking down.No practice or routine extends without end, but when the outcome is reached the demands imposed by the interaction norms subside and the situation changeseither transforming into a new one through the coming to fore of other interactive norms, or with the participants going their separate ways.

A challenge: the problem of conflicting consents
Before finishing we must address a category of cases for which this approach to consent may seem deeply problematic.An implication of the picture that we have drawn is that consent is a multi-scale, multi-layered thing, a confluence of mutual sensitivities in many different kinds of joint activity.Critical readers may be concerned that this approach implies there is consent in problematic interactions where something entirely less "shared" or "joint" obtains, particularly in situations characterised by significant asymmetries of power and privilege (Fourlas & Cuffari, 2022 explore some related themes).Moreover, the claim that a person consents simply on the basis of their being non-violent or similarly non-disruptive in a given situation is often deployed in the service of undermining marginalised individuals though the deployment and maintenance of oppressive power structures.
While it is an implication of the enactive approach sketched here that temporal chains and qualities of consent are pervasively at stake in our social interactionseven in cases where such quality seems very poor, as in subjugation and domination -we believe that this approach actually helps to articulate the complex stakes of such problematic social interactions.For such cases, a key role has to be reserved for the idea that our participation in a given interaction is a multifarious action that is being undertaken at multiple scales simultaneously.It is possible for these different facets of participation, occurring at different scales, to be in tension or conflict with one another.A person may be intentionally and willingly involved in a given setting, but nevertheless not be content with the specific manner in which the activities of that setting are unfolding.
In order to have access to the riches of social capital and material resources within a community it is necessary to be a member of that community -for instance, to be a neighbour, or to be a person of whom things can be asked -and to fulfil the responsibilities which ensure the community is sustained.A person can consent to such engagement, the broad brushstrokes of a particular social interaction, and have every right to want the particular experience or resources to which that engagement entitles them.But particular details of the interaction, the manner in which the behaviour setting is enacted in a given instance, can be problematic, and a person can not consent to this manner of interaction, resulting in a conflict of exposures to another's agency that must be negotiated.
Asymmetries of power can make such conflicts problematic, in which one person's actions make use of systemic privileges that are offensive or anathema to their fellow participants.What happens then is a difficult dance in which the joint action breaks down at one level while being maintained at another, often due to significant effort on the part of the less privileged.In these circumstances consent continues at one scale of description, while at a finer grain of analysis the privileged are being carefully managed by the marginalised (and are often oblivious to this process).
Where there are insufficient resources in the setting to enable management of the more powerful, then passivity may result, a case of being overwhelmed by a situation in which there are inadequate resources for the less privileged to negotiate in the face of power or possible violence.We are repeatedly informed by people who have suffered such violence and indecencies that the broader context, of cultural norms of victim blaming and retaliation, play a role in the apparent passivity.A person may very well enthusiastically participate in a culture and community in many respects, not least because of the coping mechanisms for certain normative exposures (professional, personal) that they are afforded.And yet that culture and community does not come about independently of violent and oppressive genres of interaction from which that person is rightly repelled, but through them.Being able to recognise and discuss these entangled dynamics of managed exposures, as joint endeavours which are sometimes in tension with each other, clearly is an important challenge.
We suggest (humbly, with an openness to correction) that the enactive approach to consent provides a means by which such problems of conflicting consents can be helpfully articulated and examined in detail.

Conclusion
Following the direction pointed to by Penny (2017Penny ( , 2022)), we have outlined an enactive approach to consent which shifts the description from one of transaction to one of manner.Within an adverbial approach, consent is a way of engaging in joint action, one in which we are sensitive not just to the level of a partner's willingness and commitment to shared direction of travel, but to its dynamics, its ebbs and flows -a dialogue more than a momentary exchange or mental state.Does this dissolve consent and make it useless as a category, indistinguishable from social interaction?No, it helps us understand why power relations are not simple, all-or-nothing things.It helps us understand what it is distinct about joint agency, and reinforces Penny's point that we must understand consent in a nuanced, continuously sensitive manner; this has received too little discussion in the philosophical literature.We believe that such an account promises resources for considering in explicit and useful terms not just the presence of consent, but its qualities and variability, in an acknowledgement of the richness and the multi-faceted nature of human joint experience.
Funding Open Access funding provided by the IReL Consortium